While the refuge concept has worked reasonably well at keeping the pests at bay, it is incompatible with eradication, Tabashnik pointed out.

"Refuges are a way of managing pest populations, so you have to accept the permanent presence of the pest," he said. "You don't get rid of them. You maintain susceptibility by promoting survival of the susceptible insects."

So, rather than relying on susceptible moths from refuges to thwart resistance, the eradication program staff reared large numbers of pink bollworms, sterilized them, and released the sterile moths into cotton fields where they could block reproduction of the wild insects.

"When a sterile moth mates with a fertile wild moth, the progeny won't be fertile," Tabashnik said. "The sterile insects soak up the reproductive potential of the wild population. If you have a high enough ratio of sterile to wild moths, you can drive the reproduction of the wild population to zero."

The release of sterile insects to control pests is not new. But this is the first time the sterile insect technique has been used in concert with a Bt crop.

A limitation of the sterile release strategy is that extremely large numbers of sterile insects can be necessary to block reproduction of the wild population, according to Tabashnik. "It becomes a question of logistics: Can you deliver enough sterile insects to overwhelm the wild population?"

"Let's say you have 1 million pests per field in non-Bt cotton fields. There is no way you could release 1 billion sterile insects per field per week."

But in combination with planting Bt cotton, the strategy turned out to be extremely powerful, Tabashnik's group discovered since tests started in 2006. Tabashnik explained: "On the Bt cotton, pink bollworm survival is virtually zero. This makes it much easier to overwhelm the wild population."

In addition to promising sustained, long term pest control without spraying insecticides for pink bollworm, the new strategy frees up acreage that otherwise would have to be set aside for refuges, thereby avoiding yield losses.

Since the eradication program began in Arizona in 2006, pink bollworm populations have declined dramatically. In 2009, only two pink bollworm larvae were found in 16,600 bolls of non-Bt cotton screened across the state. From 2005 to 2009, the pink bollworm infestation rate dropped by 99.9 percent.

Along with the decline, insecticide sprays fell to historic lows. Whereas Arizona cotton growers lost $18 million per year to pink bollworm management between 1990 and 1995, that cost plummeted to an average of $172,000 per year between 2006 and 2009.

Compared with 1995, Arizona growers' insecticide use against all cotton pests, including those not killed by Bt cotton, decreased by 88 percent, saving $200 million between 1996 and 2009.

To test the idea of delaying resistance with sterile insect releases, the researchers conducted computer simulations and analyzed more than a decade of field data from before and after deployment of this strategy statewide in Arizona.

The eradication program and associated research is a partnership among the growers, their organizations, the USDA and the UA, including Peter Ellsworth of the UA's Cooperative Extension Service.

"This has been a team effort from the get-go, with the growers essentially 100 percent on board," Tabashnik pointed out. "You can tell by the long list of authors – this wouldn't have happened without teamwork."